This article by Debra Millikan, Head Instructor of Canine Behavvioural School in Adelaide,first appeared in the New Zealand Kennel Club Gazette and then in the Canine Control Council of Queensland's 'Dog World'. It is reproduced here, with the kind permission of the author, in the hope that it will help bring about change in the way people think and how they train dogs.
In the days when people believed the earth was flat there was no evidence to indicate that it was any other way. Not until adventurers returned from the point from whence they should have fallen from the earth did people begin to comprehend that the facts did not accord with their traditional perception. Similarly, in days gone by, we had no cause to believe there was any other way of training except by the use of punishment based learning. For many years now the adventurers of dog training have been bringing back evidence that there is actually a better, kinder, more humane option for training dogs – all dogs. To make an informed choice about the methods to choose to train our dogs we need to be cognisant of the facts.
The American Veterinary Society of Animal Behavior (AVSAB) has issued a Position Statement and Guidelines regarding the use of punishment for dealing with behaviour problems in animals (www.AVSABonline.org). For the purposes of this Position Statement and Guidelines, punishment has been defined as “…the use of force, coercion or aversives to modify behavior” (AVSAB Guidelines, 2007. p. 3). The AVSAB is very clear in stating - “AVSAB’s position is that punishment should not be used as a first-line or early-use treatment for behavior problems. This is due to the potential adverse effects which include but are not limited to: inhibition of learning, increased fear-related and aggressive behaviors and injury to animals and people interacting with animals” (AVSAB Position Statement, 2007, p. 1)
The Guidelines detail both physical and psychological reasons why punishment based training is not recommended. In brief the list explains that punishment: ¨ Is difficult to time correctly ¨ Can strengthen the undesired behaviour ¨ Must be high enough in intensity for learning to occur ¨ May cause physical harm when administered at high intensity ¨ Can cause some individuals to become extremely fearful and this fear can generalise to other contexts (regardless of the strength of the punishment) ¨ Can facilitate or even cause aggressive behaviour ¨ Can suppress behaviours, including those behaviours that warn that a bite may occur ¨ Can led to a bad association ¨ Does not teach more appropriate behaviours (AVSAB Guidelines, 2007. p. 4)
The guidelines state that “Even when punishment seems mild, in order to be effective it must elicit a strong fear response…Punishment has also been shown to elicit aggressive behavior in many species of animals. Thus using punishment can put the person administering it or any person near the animal at risk of being bitten or attacked” (AVSAB Guidelines, 2007. p. 2).
Why would an organisation like the Veterinary Society of Animal Behavior feel so strongly about the use of punishment that it issues a position statement and guideline about it? Could it be that evidence based alternatives are available? The Guidelines explain that rather than punish an animal’s poor behaviour “A more appropriate approach to problem solving is to focus on reinforcing a more appropriate behavior” (AVSAB Guidelines, 2007, p. 4). This is the approach of positive reinforcement training. Not just for dogs but for all animals. What must be remembered is that basic learning theory is the same, irrespective of the animal species. Let’s face it, if you get cross with a killer whale or a sea lion which you are training, he will just swim away from you. But positive reinforcement training works for these animals (Ramirez, 1999), known for their aggression and if learning theory is the same for all animal species, why would you chose to use punishment based training on dogs?
Friedman and Brinker (2008) state “Another problem is that punishment is what most of us do best, or at least first…We are virtually surrounded by punishing strategies used to influence our behaviour: from overdue library books to dogs without licenses; fines, penalties and reprimands…For many of us, to give up punishment as our primary tool with which to influence negative behavior is to leave us empty handed”. They go on to say that it is essential that people narrow the gap between the research and practice of punishment and learn to base their choice of teaching strategies “…on facts rather than cultural inheritance”.
Extensive research on the effects of punishment has been carried out. This research “spans many decades and has been replicated with many different species of animals, including humans….the fact is there is a pattern of negative reactions or “side effects” that are consistently observed in many subjects who have been punished…” (Friedman & Brinker, 2008).
There is conflicting information in the print media, on the television and almost everywhere one seeks information. Dog guardians may feel overwhelmed, confused, conflicted and too inexperienced to choose. Sue Pearson, Certified Professional Dog Trainer with the Certification Council of Pet Dog Trainers (an international accrediting body) and holder of a Masters Degree in Education writes about training choices “If you think it might hurt the dog, it probably will. If it looks unsafe, it might be. If you wouldn’t want others to see you do it, then don’t do it. Disclaimers of “don’t try this at home” should be carefully scrutinized. “Dog friendly” training can be done safely and easily, anywhere, anytime. It can be effectively carried out by trained professionals, pet owners and even children, at home in the comfort of your living room or on the concrete drive of your local service station” (Pearson, 2006, www.iowasource.com).
So what is positive reinforcement training? “The single biggest difference between the two schools [approaches to training] is the mind set of the trainer. While the traditional trainer looks to 'correct' undesired behaviour, the positive trainer looks only for what the dog is doing right” (Bridge, 2002).
“One of the most common misconceptions about positive training is that all we do is feed the dog, regardless of what the dog is doing, or that we allow the dog to get away with bad behaviour. Nothing could be further from the truth. Positive training manipulates consequences so that the dog will do what we want.
All behaviour is driven by consequences - either the dog does something because it is reinforced (i.e. it pays off in some way, is rewarded), or the dog does something to avoid punishment or something unpleasant. Learning by consequences is a method of learning common to all species and the scientific term for this is operant conditioning.
One area of operant conditioning is called positive reinforcement. This simply means that a consequence is applied after the performance of a behaviour that makes that behaviour more likely to occur again in the future. Positive trainers take advantage of this principle when training dogs. We set the dog up so that it cannot make a mistake and then we reinforce what the dog does correctly by applying a consequence that the dog likes. Naturally the dog will want to repeat this behaviour since it gets reinforced and while he is busy doing good things he is not doing bad things. By careful management we avoid placing a dog in a situation where he can fail. This avoids the need to use punishment (which is frequently unnecessarily used in dog training) and it can hardly be considered allowing the dog to run amok.
Another method of learning common to all species is called respondent conditioning, also known as classical or Pavlovian conditioning. This is learning by association. If something is pleasant, fun and exciting, the dog will want to do it again. If an activity is boring, or scary, or associated with pain, the dog will want to avoid it. Positive trainers make training enjoyable for dogs so that they are willing partners in the training process. Positive trainers employ methods based on findings that have been described since the late 1800s, by scientists, psychologists and animal trainers, such as Edward Lee Thorndike, J.B. Watson, Ivan Pavlov, B.F. Skinner, Keller Breland, Marion-Breland Bailey, Bob Bailey, Karen Pryor and many others” (Ford, 2007. www.positivelydogs.com).
Electronic communication by the South Australian Branch of the RSPCA states: “Being positive does not mean being permissive. Our use of rewards (food, toys, play, praise and attention) is just part of a program that includes proper management to limit the dog’s opportunities to misbehave….Fortunately we do not need to rely on studies of wolf behaviour when training dogs, as there are decades of evidence-based research on the ways which all animals learn, including dogs. It’s called operant conditioning, and is the basis for our approach. Studies confirm that methods such as ours are humane, effective, and far less likely to compromise the relationship between a dog and its owner than any other [Hiby, 2004]. We agree that dogs are highly social animals that need leadership, but it is clear that we can provide this using a positive approach“ (2007. www.blog1.rspcasa.asn.au).
How to Establish Leadership without Dominance How can you tell if you are a leader of your dog or you are a follower? If your dog brings you a ball and places it on your lap, what do you do most often? Do you throw the ball for the dog? If your dog nudges you for a pat, do you pat your dog? Remember that leaders initiate and followers react! That is not to say that you should never react. Of course you can pat your dog if he asks you to. Of course you can throw the ball for him if he asks you to – just not all the time. Establish leadership by ensuring that the initiation/reaction scenario leans towards you doing more of the initiation than the dog (Ryan, n.d.). This is, in part, what positive trainers mean when they say that they manipulate the environment and the consequences to establish the behaviours that they want. This is how they set the dog up for success. This is how positive training works – manipulating the environment and the consequences.
If a strange person approaches a human-aggressive dog the dog will more than likely bark and show aggression. It is also likely that the dog’s guardian will drag the dog away or, the approaching person will retreat. Wow! The dog manipulated the consequences and made the scary person go away. Many traditional trainers would punish the dog for that aggressive behaviour. So, not only does an approaching person mean something scary to the human-aggressive dog, it also means that he is likely to get punished (hurt) each time a person approaches. This makes an approaching person even more threatening. A positive trainer will see an approaching person and alter the dog’s view of that person (manipulate the consequences of the person’s approach). This is done by desensitisation. This is “a procedure in which the dog is exposed to extremely low levels (distance, duration) of the frightening stimulus while wonderful things (the best food, toy, play, petting, praise) are happening to him (while he is calm). The level of the frightening stimulus is gradually increased but never at a rate that causes him distress” (Dennison, 2005, p56). This is the difference between the two schools of thought.
From the RSPCA (2007. www.blog1.rspcasa.asn.au) “However, for those that are determined to cling to the concept of dominance, the positive method of training is still relevant. Surely making a dog work for every bit of food that they need to live puts the owner in the most dominant position available? Leash correction training surely cannot compete with this position of power”
Despite evidence to the contrary, people persist in the belief that wolves and hence dogs have a linear hierarchy and use this to justify their continuance of dominance theory. Kathy Sdao who has a Masters Degree in Experimental Psychology and an advanced graduate degree in the science of animal behaviour, whose past career revolved around training marine mammals for the US Navy and now centres on pet dog training says “Even if dogs did form linear packs, there’s no evidence to suggest that they perceive humans as part of their species specific ranking. In general, humans lack the capability to even recognize, let alone replicate, the elegant subtleties of canine body language. So it’s hard to imagine that dogs could perceive us as pack members at all” (Sdao, 2007. www.dogcentral.msn.com). A well rounded program that teaches using positive reinforcement also includes training in understanding elements of dog body language. A dog guardian’s ability to comprehend what his dog is “saying” with his body is more valuable than oil! If each pet guardian can be taught to understand some of those subtle signals and learn how to correctly address some of the issues his dog is “talking” about, there would be far fewer aggressive dogs in this world. Our ability to comprehend at least in part what our dogs are telling us is yet another example of how positive trainers set up scenarios of success.
Don’t wait until it is too late and the need for correction ever present – manipulate the environment and the consequences of the dog’s actions, based on knowledge of the dog and his basic needs and individual idiosyncrasies and you have a recipe for success. Training with aversives is like getting into a taxi and telling the driver all the places you don’t want to go. We need to get into the taxi of dog training with the address of the exact point we want to go. If we do, the ride will be less harrowing for the guardian and the destination for the dog a more secure and trusting one.
Having spent twenty of my thirty years of dog training using punishment based methods, I can speak from experience about the benefits of positive reinforcement training. Having crossed over from traditional methods to positives I could never cross back. Correctly understood, taught and undertaken, positive training is fun and can lead dogs and their guardians into a wonderful life-long relationship built on consistency, trust and respect, within clearly defined guidelines. Yes, dog guardians do need to make informed decisions that are very relevant to their entire family and their dogs. That is why it is very important that training methods are presented with a full knowledge and understanding of each. After all, in any learning scenario, presentation of fact, not what is perceived to be fact, aids both teacher and student.
Debra Millikan, PDT Professional Dog Trainer, Association of Animal Behavior Professionals www.associationofanimalbehaviorprofessionals.com Cert IV Dog Behavioural Training Delta Accredited Canine Good Citizen Instructor www.deltasocietyaustralia.com
Bibliography ¨ AVSAB Guidelines on the Use of Punishment for Dealing with Behavior Problems in Animals. 2007. Retrieved on 18 July, 2008 from http://www.vetmed.ucdavis.edu/small_animal/services/behavior/AVSAB_Punishment_Statements.pdf ¨ AVSAB Position Statement on the Use of Punishment for Behavior Modification in Animals. 2007. Retrieved on 18 July, 2008 from www.vetmed.ucdavis.edu/small_animal/services/behavior/AVSAB_Punishment_Statements.pdf ¨ Bridge K. 2002. Positive Training: What it is and what it is not. NSW Canine Journal ¨ Dennison, P, 2005. How to Right a Dog Gone Wrong. Alpine Blue Ribbon Books. Colorado, USA ¨ Ford, L. 2007. What is Positive Training? Retrieved on 18 July, 2008 from www.positivelydogs.com ¨ Friedman, SG & Brinker B 2008. The facts about punishment. Journal of Applied Companion Animal Behavior 2(1), 12-15. ¨ Hiby EF, Rooney NJ, Bradshaw JWS. Dog Training Methods: Their Use, Effectiveness and Interaction with Behaviour and Welfare. Animal Welfare 2004, 13:63-69. ¨ Ramirez, K. 1999. Animal Training: Successful Animal Management Through Positive Reinforcement. Shed Aquarium Society, Chicago, USA ¨ RSPCA 2007. Positive reinforcement versus punishment in dog training. Retrieved on 18 July, 2008 from www.blog1.rspcasa.asn.au/2007/03/08/positive-reinforcement-versus-punishment-in-dog-training/ ¨ Ryan K. (n.d.) Are You the Alpha? Retrieved on 21 July, 2008 from www.4pawsu.com/leaderadv.htm ¨ Sdao, K. 2007. Forget About Being Alpha in Your Pack. Retrieved on 21 July, 2008 from www.dogcentral.msn.com/article.aspx?cp-documentid=4936120
I truly believe that this country, and maybe the world, has gotten so involved in being politically correct, they have lost sight of everything tried and true. Does anyone remember how the school bully was handled ? They got their a$$ handed to them, eventually, and guess what? The person being picked on wasn't picked on anymore. God forbid you tell your child anymore that it is OK to stand up for themselves, and put a bully in their place. Kids are taught in kindergarten to never use violence against anyone for any reason. So the kids grow up being more afraid of getting in trouble with the school officials and their parents for getting in a fight, instead of knowing how to properly deal with and resolve conflict. Violence is NOT the answer to every conflict, but it does have its place. I believe that was the entire gist of the previous thread. My personal belief is that once an animal has been taught what is and is not acceptable, there is nothing wrong with physical correction.
If a puppy is snarling and hanging off my arm, I am not going to run for it's favorite toy to do a pleasant exchange. Maybe I am wrong on this, but following the theory of operant conditioning, the puppy then learns that if he bites my arm, he gets his favorite toy. I am probably off on that but it does seem to me that the puppy would be rewarded instead of being taught that hanging off my arm is a serious offense and one that is never to be repeated.
Again I have to reiterate that not all training methods work for all dogs. Years of experience, dealing with multiple types of personalities, has taught me that much.
Now if someone would point me in the right direction on how to train a cat I'd be appreciative. I am sure Tart would be too, before he finds his butt permanently outdoors !!
Scout, its not that you are rewarding a pup for biting. Biting is normal puppy behaviour. Along with teaching bite inhibition I redirect him with a toy he can chew on, show him what I want instead of punishment.
Personally I think the positive reinforcement and no punishment training is for the birds. I do not believe you should beat your animal or kid, but I do believe that they need to know what they did was wrong. Not all training works for all people, animals, etc. but I have never had a dog have any ill effects from punishment used correctly. I also firmly believe that kids in today society would not be so out of control if they got their hands/butts smacked a few times as kids. If my mother had ever tried putting me in a time out, I can tell you it would have never worked.
I see way to many ill trained animals and children. Everyone has a right to thier own opinion. As far as I see it, I will keep my old fashioned way of training. It has worked for me for years, and I do not a reason to "fix" something that is not broke.
I had an instructor at petsmart try pulling the treat over the nose trick on one of my aussies. All she did was back up to a wall and stand there. When the wall turned she turned with the wall. The instructor looked at me and told me I needed to work on my sits. If she would have asked my dog to sit, she would have. I train them by showing them, not waiting until they figure it out. My aussies don't work like that.
I use simular ideals with my dogs and my kids. If they do something right, they get praise and possibly a treat. If they do something wrong, they get punnished.
I have told my kids when it comes to fights, don't start it, finnish it. My daughter is 10, in 4th grade, and hasn't had a bully pick on her since 1st grade, when punched a kid who pushed her around. My son is 6, in kindergarten, and is yet to expierence it. However, when he was about 3, he saw someone picking on his older sister and told them "You don't pick on my sister" and then punched the kid in the stomach, hard. I saw this and had to walk a fine line. I told him "You shouldn't hit. That being said, well done for sticking up for your sister."
Too many people have ill mannered pets and children. I think the problems are the same: lack of structure, dicipline, and not properly using their energy.
Beautiful new babies. Mama and all are doing well.
I agree with most everyone else who has posted. I don't see how giving a dog who has been taught not to bite a treat when he does will solve the problem. It will only lead to more problems. Scout hit the nail on the head its operant conditioning and the wrong kind too. I have four dogs two that are two years old and two less than six months. My older two got popped in the mouth when they bit as puppies and its never been a problem since, I'm doing it the same way with my youngest two because why change what works. I disagree that biting is NORMAL puppy behavior. Nipping maybe and chewing definitely those two things I correct by giving toys...but a bite? Pop to the mouth for sure.
If your dog doesn't like someone you probably shouldn't either!
Can we keep the topic on dog punishment training? Punishment based training can work but would you rather have a dog that works out of avoidence or one that is happy and eager to learn? What do you think hitting a puppy or dog actualy teaches him?
Deputydog, did you have something to add to the discussion?
I admit that only skimmed the article, but here's my two cents. At least half the items in the list of possible negative effects looks like they could just as easily apply to positive reinforcement (e.g. bad timing, incorrect associations).
If you consider punishment and rewards as simply tools in a tool box, I think we can be closer to agreeing on this issue. You use the tool that works for the situation. If treats, praise, and redirection are working, you use them. I think what people are saying is there are times when a different tool does the job faster, and more effectively. After all, you can use a hammer to pound in a screw, but it might be faster and better for everyone if you use a screwdriver instead.
***Edited By: NoDogYet on 5/12/2010 11:45:41 AM*** Reason: sp
Well my mistake, I thought we were talking about that. My dogs don't avoid me, quite the opposite actually they follow me everywhere I go. None of my dogs cower when I come near, in fact they wag their tails. In some cases punishment is necessary, giving treats and praise and toys doesn't always work. When you were younger, had you done something like say, skip school, would your mother and father have said "Oh how wonderful, here have some candy and whatever else you like." No I don't think so. When I got caught doing something I shouldn't have done, I got punished plain and simple. And you know what, IT WORKED. My dogs listen and they're all happy dogs and THEY'RE ALL EAGER TO PLEASE.
What it really comes down to here is that you're a little irritated that no one seems to be agreeing with what you say.
If your dog doesn't like someone you probably shouldn't either!
I posted those smiles because I find it amusingly ironic that so far the only responses people have made have been in defense of their position on the use of "Physical Corrective Action".
I did read your article just as I read every post you made in the previous thread. There are many good points in this article and there are some points I do not agree with. The reason I won't debate the points I don't agree with is because I really don't want to rehash old arguments. Our positions are not going to change.
You and Catlover have both offered some good training tips. Many methods that I myself use. My point all along has been that "Physical Corrective Action" has it's place within that training program.
They're are things that you and I will not see eye to eye on. What I really don't understand is why you seem to be trying so hard to convince those of us that know that this IS an effective training tool in certain difficult situations, that it does not work.
You can't win if you don't play the game -- Me!
All cruelty stems from weakness -- Seneca
Sometimes a dog is as good as any man -- Bell/Houser
Catlover, I don't think that's at all what he's saying. I think he's saying make valid points otherwise you are inferior to the conversation. People can agree to disagree, when you make logical arguments.
If your dog doesn't like someone you probably shouldn't either!
I have read the article, again, and one thing I fail to notice is what do these postive reinforcement trainers do when they encounter a behavior that is undesirable? There are both pleasant and unpleasant consequences for everything in life. I pay my electric bill, my lights stay on. I don't pay it, lights out. Of course having the lights stay on would be my "positive" incentive to pay my bill. But suppose 1 month I decide to not pay the bill? Guess what...lights go out. I learn to make more of an effort to make sure that bill gets paid the following month.
I still fail to see how using positive reinforcement to TRAIN your pup, and negative reinforcement to correct undesirable behavior is so wrong.
Maybe the confusion is in your following statement:
Punishment based training can work but would you rather have a dog that works out of avoidence or one that is happy and eager to learn? What do you think hitting a puppy or dog actualy teaches him?
To me the word "works" implies the dog has already learned a command. In which case yes, if the dog fails to respond to said command there has to be a negative consequence for disobedience. A "happy and eager to learn dog" implies one that has not yet learned what is expected of him. In which case, learning must occur first before any type of consequence can be applied for disobedience. You have taken 2 stages of the learning process, the beginning and the end, and have combined them to mean only 1 stage.
I also disagree wholeheartedly with the following paragraph:
How can you tell if you are a leader of your dog or you are a follower? If your dog brings you a ball and places it on your lap, what do you do most often? Do you throw the ball for the dog? If your dog nudges you for a pat, do you pat your dog? Remember that leaders initiate and followers react! That is not to say that you should never react. Of course you can pat your dog if he asks you to. Of course you can throw the ball for him if he asks you to – just not all the time. Establish leadership by ensuring that the initiation/reaction scenario leans towards you doing more of the initiation than the dog (Ryan, n.d.). This is, in part, what positive trainers mean when they say that they manipulate the environment and the consequences to establish the behaviours that they want. This is how they set the dog up for success. This is how positive training works – manipulating the environment and the consequences.
My dogs will bring toys over wanting to play, they will come over and solicit petting constantly. I have never ignored a request for either and I never will. When I have tired of petting them or playing with them, I say enough and they know playtime, snuggling time is over. They walk away happy babies, not leaders. Which brings me to the next issue I did not see written in this particular article but is a favorite amongst those who train this way ... the dreaded TUG-OF-WAR game. I at first listened to the trainers who said you must always end the tug of war game and never let your dog win in order to maintain leadership. Know what I found out? Dogs get really bored very quickly with a game they are never allowed to win. You want your dog to never play another game of tug? Then never let them win and they will lose interest in that game very quickly. So now whenever I play tug, we each win a round and now they will play tug with me forever, if I would play it forever with them lol. I have seen this happen over and over with my son and his dog. My son was also under the impression he always had to win. Max stopped playing very quickly when he wasn't winning. So I asked my son how often he would play a game if he could never win at it? He said he would lose interest after awhile and probably never play anymore. So I made him see the relationship between how he felt and Max walking away from the game. Now he allows Max to win a few rounds and they have FUN. My son ends the game when he's had enough and both of them walk away from playing happy.
Corsomom, You do realize that the article states: The AVSAB is very clear in stating - “AVSAB’s position is that punishment should not be used as a first-line or early-use treatment for behavior problems.
Meaning...this shouldn't be the first method you use? That's not stating that punishment is a method that should NEVER be used. When I first got my pups I tried "No." I tried toys...and it didn't work...so I used the pop to the mouth. The very beginning of the article states NOT TO BE USED AS A FIRST-LINE OR EARLY-USE TREATMENT...it doesn't say...SHOULD NEVER EVER BE USED. Popping a dog in the mouth when they bite will teach them NOT to do the behavior they were doing because it's unwanted...PLAIN AND SIMPLE. It's not going to teach the dog to fear you...that will only happen if you BEAT your dog. As I said, pops to the mouth has ALWAYS worked for me, and you know...my parents did it with their dogs and they're well behaved dogs, as well as everyone else in my family. I'd also like to point out...NOT ONE DOG IN THE FAMILY fears their owner, and THEY'RE ALL well behaved dogs.
If your dog doesn't like someone you probably shouldn't either!
"What it really comes down to here is that you're a little irritated that no one seems to be agreeing with what you say."
I'm not irritated if someome does not agree with me. I just think if someone ones to talk about children and punishment it belongs in another thread, probably in general.
Hopefuly we can keep this thread peaceful.
I was watching video's yesterday of dogs that were being trained using punishment and dogs being trained using positive reinforcement. The difference in the dogs body language was so obvious. I'll try and post some of the video's later.
Instructing owners to eat before their dog or go through doors first will not influence the dog’s overall perception of the relationship – merely teach them what to expect in these specific situations.
Thank you. I have said that over and over and yet people still seem to believe in that malarkey.